Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “hyperparameter optimization”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 199 records · Page 11

Using Flory–Huggins-informed human-in-the-loop Bayesian optimization to map the phase diagram of polymer blends

Mapping the phase diagram of polymer blends is an essential step in controlling the structure–property relationship of polymer-based materials. However, traditional grid-based approaches are inefficient and rely on subjective judgements for terminating the experimental campaign. Artificial intelligence-guided experimentation offers a compelling alternative, especially when data-driven decision-making is interfaced with established polymer thermodynamics to improve efficiency and interpretability. Here, we introduce a physics-informed Bayesian optimization approach to guide the mapping of the phase diagram of a model blend containing poly(methyl methacrylate) and poly(styrene-ran-acrylonitrile). Physical information is derived from a Flory–Huggins representation of the spinodal curve, which is integrated into the Bayesian optimization process as a structured prior mean that acts as a soft constraint. Implemented as a human-in-the-loop workflow, the approach leverages optical imaging of film cloudiness with iterative Gaussian process surrogate modeling and a parameter selection decision policy to identify the composition-temperature conditions for sequential iterations. Convergence of kernel and Flory–Huggins-based hyperparameters provided a stopping criterion, ensuring an objective and interpretable termination of the experimental campaign. The framework recovered the known lower critical solution temperature (∼160 °C), while increasing material efficiency through targeted sampling. This work establishes a proof-of-concept for the application of Bayesian optimization workflows to study polymer blend miscibility.

36 MATERIALS SCIENCE↗

Improving Deep Neural Networks’ Training for Image Classification With Nonlinear Conjugate Gradient-Style Adaptive Momentum

Momentum is crucial in stochastic gradient-based optimization algorithms for accelerating or improving training deep neural networks (DNNs). In deep learning practice, the momentum is usually weighted by a well-calibrated constant. However, tuning the hyperparameter for momentum can be a significant computational burden. In this article, we propose a novel adaptive momentum for improving DNNs training; this adaptive momentum, with no momentum-related hyperparame- ter required, is motivated by the nonlinear conjugate gradient (NCG) method. Stochastic gradient descent (SGD) with this new adaptive momentum eliminates the need for the momentum hyperparameter calibration, allows using a significantly larger learning rate, accelerates DNN training, and improves the final accuracy and robustness of the trained DNNs. For example, SGD with this adaptive momentum reduces classification errors for training ResNet110 for CIFAR10 and CIFAR100 from 5.25% to 4.64% and 23.75% to 20.03%, respectively. Furthermore, SGD, with the new adaptive momentum, also benefits adversarial training and, hence, improves the adversarial robustness of the trained DNNs.

97 MATHEMATICS AND COMPUTING↗

IL-Net: Using Expert Knowledge to Guide the Design of Furcated Neural Networks

Deep neural networks (DNN) excel at extracting patterns. Through representation learning and automated feature engineering on large datasets, such models have been highly successful in computer vision and natural language applications. Designing optimal network architectures from a principled or rational approach however has been less than successful, with the best successful approaches utilizing an additional machine learning algorithm to tune the network hyperparameters. However, in many technical fields, there exist established domain knowledge and understanding about the subject matter. In this work, we develop a novel furcated neural network architecture that utilizes domain knowledge as high-level design principles of the network. We demonstrate proof-of-concept by developing IL-Net, a furcated network for predicting the properties of ionic liquids, which is a class of complex multi-chemicals entities. Compared to existing state-of-the-art approaches, we show that furcated networks can improve model accuracy by approximately 20-35%, without using additional labeled data. Lastly, we distill two key design principles for furcated networks that can be adapted to other domains.

Sakloth, Khushmeen↗

Fully Convolutional Spatio-Temporal Models for Representation Learning in Plasma Science

We have trained a fully convolutional spatio-temporal model for fast and accurate representation learning in the challenging exemplar application area of fusion energy plasma science. The onset of major disruptions is a critically important fusion energy science issue that must be resolved for advanced tokamak plasmas such as the $25B burning plasma international thermonuclear experimental reactor (ITER) experiment. While a variety of statistical methods have been used to address the problem of tokamak disruption prediction and control, recent approaches based on deep learning have proven particularly compelling. In the present paper, we introduce further improvements to the fusion recurrent neural network (FRNN) software suite, which delivered cross-machine disruption predictions with unprecedented accuracy using a large database of experimental signals from two major tokamaks. Up to now, FRNN was based on the long short-term memory (LSTM) variant of recurrent neural networks to leverage the temporal information in the data. Here, we implement and apply the "temporal convolutional neural network (TCN)" architecture to the time-dependent input signals. Furthermore, this allows highly optimized convolution operations to carry the majority of the computational load of training, thus enabling a reduction in training time, and the effective use of high-performance computing resources for hyperparameter tuning. At the same time, the TCN-based architecture achieves better predictive performance when compared with the LSTM architecture for various tasks for a representative fusion database.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Predicting initial trans-membrane pressure across cycles in the ultrafiltration process using random forest

With growing freshwater scarcity, direct potable reuse (DPR) systems that reclaim wastewater for drinking are becoming increasingly important for sustainable water supply. Reliable operation requires minimizing downtime in ultrafiltration (UF) units, where membrane fouling leads to elevated trans-membrane pressure (TMP). This study develops data-driven regression models based on random forest (RF) and autoregressive (AR) approaches to forecast the initial TMP at the start of each UF filtration cycle in a pilot-scale DPR system. The RF model consistently outperforms baseline methods, including historical mean, last observation carried forward, and AR models, across multiple forecast horizons, achieving the lowest root mean square error. To evaluate how different classes of process variables contribute to TMP dynamics over time, we examine the feature importance of independent input variables across multiple forecast horizons. This analysis provides insight into the temporal relevance of operational and sensor-derived features, guiding control and monitoring strategies. Additionally, the impact of hyperparameter tuning on TMP prediction performance is assessed for both direct and recursive RF modelling approaches. The proposed RF framework establishes a robust foundation for predictive monitoring and real-time optimization of UF operations, supporting sustainable and reliable water reuse.

direct potable reuse↗

Use of Design of Experiments and Rule-Based Inference in Determining Neural Network Architectures for Loss of Control Detection

In this work, we describe methods for selecting the neural network architectures and input spaces to implement belief state inference on generic commercial transport aircraft. First, we highlight a case study on the planning, execution, and analysis of a set of experiments to determine the configurations of a conditional variational autoencoder (CVAE). We present a structured method that can be used in a number of aerospace applications, to optimize the structure and training parameters of the CVAE for belief state inference, using Design of Experiments (DOE) statistical methodologies. The motivation for this specific DOE was to identify the appropriate hyperparameters for measuring the CVAE reconstruction probability and latent space, such that the measurements can be used to infer qualitative state changes for the aircraft. We demonstrate that this process yields information about a trained neural network’s utility for this specific application, along with a quantifiable range of certainty. We execute 84 experiments using loss-of-control flight maneuver data from a NASA T-2 aircraft, demonstrating that this empirical process allows us to construct cheap and simple models with specific attributes amenable to belief state inference in aerospace applications. While theoretically, we could create a single CVAE with an input space the size of all measurable flight variables and environmental dynamics, it becomes intractable to use such a neural network in an in-situ intelligent multi-agent system. Using the recommendations from our case study, we introduce a technical approach for feasibly describing the belief space by (1) identifying significant statistical relationships among flight variables using rule induction, (2) using a set of rules that cover all features to define the input space of multiple CVAEs, and (3) forming a belief space based on the joint probability density of their collective latent spaces. This results in a series of relatively small matrix multiplications that can be performed in real time, as opposed to large matrix computations in a single CVAE. We demonstrate the application of this approach on the T-2 flight loss-of control experiments, using the architecture and hyperparameter recommendations from the case study. We compare the utilities of an individual CVAE trained on all flight variables and multiple CVAEs defined on subsets of flight variables for detecting qualitative changes in flight. We demonstrate that the use of multiple CVAEs with smaller input spaces permits the CVAE to capture more granular relationships in the latent space, permitting better state space characterization and loss-of-control detection.

Design of experiments↗

Deep Learning for Modeling Enhanced Geothermal Systems

Enhanced Geothermal Systems (EGS) offer a vast potential to expand the use of geothermal energy. Heat is extracted from this engineered system by injecting cold water into a subsurface fractures, which are in contact with the hot dry rock, and pulled through the production wells. Creating EGS requires improving the natural permeability of hot crystalline rocks. To develop economically viable EGS reservoirs, significant technical barriers (e.g., better stimulation technologies without adequate water and/or permeability) and non-technical barriers (e.g., land access and permitting) must be overcome. In this short conference paper, we present a workflow to address a part of this challenge – “How to develop economically viable EGS using existing technologies?”. Our workflow called the GeoThermalCloud (GTC) for EGS, leverages recent advances in machine learning, deep learning, and cloud computing. This GTC framework is open-source and available at https://github.com/SmartTensors/GeoThermalCloud.jl. The GTC framework provides trained deep learning (DL) models to estimate the net present value of a given EGS design scenario. The Geothermal Design Tool (https://github.com/GeoDesignTool/GeoDT.git), a fast and simplified multi-physics solver, is used to develop a database for training DL models. The database consists of EGS design parameters (inputs to DL model) and their net present value (output of DL model) in uncertain geologic systems. The EGS design parameters for constructing this training database are based on Utah FORGE but include the options of more wells and deeper depths. The DL models are trained by ingesting the EGS design parameters and estimating the corresponding net present value. Such an emulation allows us to screen various EGS designs quickly and identify good development strategies by coupling them with optimization techniques. Our preliminary results show promise in DL emulation of net present value. However, a lot more work is needed to improve the predictive capability of DL models (i.e., extensive hyperparameter tuning is necessary). This will be the primary focus of our future work.

artificial neural networks, geothermal↗

Data-free inference of the joint distribution of uncertain model parameters.

It is known that, in general, the correlation structure in the joint distribution of model parameters is critical to the uncertainty analysis of that model. Very often, however, studies in the literature only report nominal values for parameters inferred from data, along with confidence intervals for these parameters, but no details on the correlation or full joint distribution of these parameters. When neither posterior nor data are available, but only summary statistics such as nominal values and confidence intervals, a joint PDF must be chosen. Given the summary statistics it may not be reasonable nor necessary to assume the parameters are independent random variables. We demonstrate, using a Bayesian inference procedure, how to construct a posterior density for the parameters exhibiting self consistent correlations, in the absence of data, given (1) the fit-model, (2) nominal parameter values, (3) bounds on the parameters, and (4) a postulated statistical model, around the fit-model, for the missing data. Our approach ensures external Bayesian updating while marginalizing over possible data realizations. We then address the matching of given parameter bounds through the choice of hyperparameters, which are introduced in postulating the statistical model, but are not given nominal values. We discuss some possible approaches, including (1) inferring them in a separate Bayesian inference loop and (2) optimization. We also perform an empirical evaluation of the algorithm showing the posterior obtained with this data free inference compares well with the true posterior obtained from inference against the full data set.

99 GENERAL AND MISCELLANEOUS↗

Data-free inference of uncertain model parameters.

It is known that, in general, the correlation structure in the joint distribution of model parameters is critical to the uncertainty analysis of that model. Very often, however, studies in the literature only report nominal values for parameters inferred from data, along with confidence intervals for these parameters, but no details on the correlation or full joint distribution of these parameters. When neither posterior nor data are available, but only summary statistics such as nominal values and confidence intervals, a joint PDF must be chosen. Given the summary statistics it may not be reasonable nor necessary to assume the parameters are independent random variables. We demonstrate, using a Bayesian inference procedure, how to construct a posterior density for the parameters exhibiting self consistent correlations, in the absence of data, given (1) the fit-model, (2) nominal parameter values, (3) bounds on the parameters, and (4) a postulated statistical model, around the fit-model, for the missing data. Our approach ensures external Bayesian updating while marginalizing over possible data realizations. We then address the matching of given parameter bounds through the choice of hyperparameters, which are introduced in postulating the statistical model, but are not given nominal values. We discuss some possible approaches, including (1) inferring them in a separate Bayesian inference loop and (2) optimization. We also perform an empirical evaluation of the algorithm showing the posterior obtained with this data free inference compares well with the true posterior obtained from inference against the full data set.

99 GENERAL AND MISCELLANEOUS↗

Use of Design of Experiments in Determining Neural Network Architectures for Loss of Control Detection

We describe empirical methods for selecting a neural network architecture to implement belief state inference on generic commercial transport aircraft. We highlight a case study on the planning, execution, and analysis of a set of experiments to determine the configurations of a conditional variational autoencoder (CVAE). Our main contribution is the application of a structured method that can be used for machine learning in many aerospace applications. This method optimizes the structure and training parameters of a neural network for belief state inference, using Design of Experiments (DOE) statistical methodologies. The motivation for this specific DOE analysis was to identify the appropriate hyperparameters for measuring the CVAE reconstruction probability and latent space, such that the measurements can be used to infer qualitative state changes for the aircraft. We demonstrate that this process yields information about a trained neural network’s utility for this specific application, along with a quantifiable range of certainty. We execute 84 experiments using loss-of-control flight maneuver data from the NASA T-2 aircraft, demonstrating that this empirical process allows us to construct cheap and simple models with specific attributes amenable to belief state inference in aerospace applications.

Loss of Control↗

Use of Design of Experiments in Determining Neural Network Architectures for Loss of Control Detection

We describe empirical methods for selecting a neural network architecture to implement belief state inference on generic commercial transport aircraft. We highlight a case study on the planning, execution, and analysis of a set of experiments to determine the configurations of a conditional variational autoencoder (CVAE). Our main contribution is the application of a structured method that can be used for machine learning in many aerospace applications. This method optimizes the structure and training parameters of a neural network for belief state inference, using Design of Experiments (DOE) statistical methodologies. The motivation for this specific DOE analysis was to identify the appropriate hyperparameters for measuring the CVAE reconstruction probability and latent space, such that the measurements can be used to infer qualitative state changes for the aircraft. We demonstrate that this process yields information about a trained neural network’s utility for this specific application, along with a quantifiable range of certainty. We execute 84 experiments using loss-of-control flight maneuver data from the NASA T 2 aircraft, demonstrating that this empirical process allows us to construct cheap and simple models with specific attributes amenable to belief state inference in aerospace applications.

Loss of Control↗

Use of Design of Experiments in Determining Neural Network Architectures for Loss of Control Detection

Abstract—We describe empirical methods for selecting a neural network architecture to implement belief state inference on generic commercial transport aircraft. We highlight a case study on the planning, execution, and analysis of a set of experiments to determine the configurations of a conditional variational autoencoder (CVAE). Our main contribution is the application of a structured method that can be used for machine learning in many aerospace applications. This method optimizes the structure and training parameters of a neural network for belief state inference, using Design of Experiments (DOE) statistical methodologies. The motivation for this specific DOE analysis was to identify the appropriate hyperparameters for measuring the CVAE reconstruction probability and latent space, such that the measurements can be used to infer qualitative state changes for the aircraft. We demonstrate that this process yields information about a trained neural network’s utility for this specific application, along with a quantifiable range of certainty. We execute 84 experiments using loss-of-control flight maneuver data from the NASA T-2 aircraft, demonstrating that this empirical process allows us to construct cheap and simple models with specific attributes amenable to belief state inference in aerospace applications.

neural networks↗

Optimizing Classifiers for Radionuclide Identification

Identifying threat nuclear materials is a critical for the prevention of acts of nuclear terrorism on the homeland. For this purpose, many radionuclide identification devices are deployed in the field. However, these will not necessarily be in the hands of non-experts, therefore these devices need to provide ready-made answers for the personnel in the field. This is where advanced algorithms are employed to both interpret the data and provide the identification of the nuclear material being interrogated. We took a machine learning approach to identification, by using training and validation data sets to create and optimize classifiers which determine which radionuclide is consistent with the data. The classifiers investigated were the Random Forest, Decision Tree, Support Vector Machine, and XG Boost and their performance was judged using the F1 Score for both hyperparameter tuning and comparison. In the end, we found out that the Random Forest Classifier worked the best based off the F1 Score they got which was 0.98.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

Effective noise-suppressed and artifact-reduced reconstruction of SPECT data using a preconditioned alternating projection algorithm

Purpose: The authors have recently developed a preconditioned alternating projection algorithm (PAPA) with total variation (TV) regularizer for solving the penalized-likelihood optimization model for single-photon emission computed tomography (SPECT) reconstruction. This algorithm belongs to a novel class of fixed-point proximity methods. The goal of this work is to investigate how PAPA performs while dealing with realistic noisy SPECT data, to compare its performance with more conventional methods, and to address issues with TV artifacts by proposing a novel form of the algorithm invoking high-order TV regularization, denoted as HOTV-PAPA, which has been explored and studied extensively in the present work. Methods: Using Monte Carlo methods, the authors simulate noisy SPECT data from two water cylinders; one contains lumpy “warm” background and “hot” lesions of various sizes with Gaussian activity distribution, and the other is a reference cylinder without hot lesions. The authors study the performance of HOTV-PAPA and compare it with PAPA using first-order TV regularization (TV-PAPA), the Panin–Zeng–Gullberg one-step-late method with TV regularization (TV-OSL), and an expectation–maximization algorithm with Gaussian postfilter (GPF-EM). The authors select penalty-weights (hyperparameters) by qualitatively balancing the trade-off between resolution and image noise separately for TV-PAPA and TV-OSL. However, the authors arrived at the same penalty-weight value for both of them. The authors set the first penalty-weight in HOTV-PAPA equal to the optimal penalty-weight found for TV-PAPA. The second penalty-weight needed for HOTV-PAPA is tuned by balancing resolution and the severity of staircase artifacts. The authors adjust the Gaussian postfilter to approximately match the local point spread function of GPF-EM and HOTV-PAPA. The authors examine hot lesion detectability, study local spatial resolution, analyze background noise properties, estimate mean square errors (MSEs), and report the convergence speed and computation time. Results: HOTV-PAPA yields the best signal-to-noise ratio, followed by TV-PAPA and TV-OSL/GPF-EM. The local spatial resolution of HOTV-PAPA is somewhat worse than that of TV-PAPA and TV-OSL. Images reconstructed using HOTV-PAPA have the lowest local noise power spectrum (LNPS) amplitudes, followed by TV-PAPA, TV-OSL, and GPF-EM. The LNPS peak of GPF-EM is shifted toward higher spatial frequencies than those for the three other methods. The PAPA-type methods exhibit much lower ensemble noise, ensemble voxel variance, and image roughness. HOTV-PAPA performs best in these categories. Whereas images reconstructed using both TV-PAPA and TV-OSL are degraded by severe staircase artifacts; HOTV-PAPA substantially reduces such artifacts. It also converges faster than the other three methods and exhibits the lowest overall reconstruction error level, as measured by MSE. Conclusions: For high-noise simulated SPECT data, HOTV-PAPA outperforms TV-PAPA, GPF-EM, and TV-OSL in terms of hot lesion detectability, noise suppression, MSE, and computational efficiency. Unlike TV-PAPA and TV-OSL, HOTV-PAPA does not create sizable staircase artifacts. Moreover, HOTV-PAPA effectively suppresses noise, with only limited loss of local spatial resolution. Of the four methods, HOTV-PAPA shows the best lesion detectability, thanks to its superior noise suppression. HOTV-PAPA shows promise for clinically useful reconstructions of low-dose SPECT data.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Approximate Bayesian Model Inversion for PDEs with Heterogeneous and State-Dependent Coefficients

We present two approximate Bayesian inference methods for parameter estimation in partial differential equation (PDE) models with space-dependent and state-dependent parameters. We demonstrate that these methods provide accurate and cost-effective alternatives to Markov Chain Monte Carlo simulation. We assume a parameterized Gaussian prior on the unknown functions, and approximate the posterior density by a parameterized multivariate Gaussian density. The parameters of the prior and posterior are estimated from sparse observations of the PDE model's states and the unknown functions themselves by maximizing the evidence lower bound (ELBO), a lower bound on the log marginal likelihood of the observations. The first method, Laplace-EM, employs the expectation maximization algorithm to maximize the ELBO, with a Laplace approximation of the posterior on the E-step, and minimization of a Kullback-Leibler divergence on the M-step. The second method, DSVI-EB, employs the doubly stochastic variational inference (DSVI) algorithm, in which the ELBO is maximized via gradient-based stochastic optimization, with nosiy gradients computed via simple Monte Carlo sampling and Gaussian backpropagation. We apply these methods to identifying diffusion coefficients in linear and nonlinear diffusion equations, and we find that both methods provide accurate estimates of posterior densities and the hyperparameters of Gaussian priors. While the Laplace-EM method is more accurate, it requires computing Hessians of the physics model. The DSVI-EB method is found to be less accurate but only requires gradients of the physics model.

approximate Bayesian inference, model inversion, v↗

Scaling the training of particle classification on simulated MicroBooNE events to multiple GPUs

Measurements in Liquid Argon Time Projection Chamber (LArTPC) neutrino detectors, such as the MicroBooNE detector at Fermilab, feature large, high fidelity event images. Deep learning techniques have been extremely successful in classification tasks of photographs, but their application to LArTPC event images is challenging, due to the large size of the events. Events in these detectors are typically two orders of magnitude larger than images found in classical challenges, like recognition of handwritten digits contained in the MNIST database or object recognition in the ImageNet database. Ideally, training would occur on many instances of the entire event data, instead of many instances of cropped regions of interest from the event data. However, such efforts lead to extremely long training cycles, which slow down the exploration of new network architectures and hyperparameter scans to improve the classification performance. We present studies of scaling a LArTPC classification problem on multiple architectures, spanning multiple nodes. The studies are carried out on simulated events in the MicroBooNE detector. We emphasize that it is beyond the scope of this study to optimize networks or extract the physics from any results here. Institutional computing at Pacific Northwest National Laboratory and the SummitDev machine at Oak Ridge National Laboratory’s Leadership Computing Facility have been used. To our knowledge, this is the first use of state-of-the-art Convolutional Neural Networks for particle physics and their attendant compute techniques onto the DOE Leadership Class Facilities. We expect benefits to accrue particularly to the Deep Underground Neutrino Experiment (DUNE) LArTPC program, the flagship US High Energy Physics (HEP) program for the coming decades.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Monte Carlo Tree Search Methods for the Earth-Observing Satellite Scheduling Problem

This work explores on-board planning for the single spacecraft, multiple ground station Earth-observing satellite scheduling problem through artificial neural network function approximation of state–action value estimates generated by Monte Carlo tree search (MCTS). An extensive hyperparameter search is conducted for MCTS on the basis of performance, safety, and downlink opportunity utilization to determine the best hyperparameter combination for data generation. A hyperparameter search is also conducted on neural network architectures. The learned behavior of each network is explored, and each network architecture’s robustness to orbits and epochs outside of the training distributions is investigated. Furthermore, each algorithm is compared with a genetic algorithm, which serves to provide a baseline for optimality. MCTS is shown to compute near-optimal solutions in comparison to the genetic algorithm. The state–action value networks are shown to match or exceed the performance of MCTS in six orders of magnitude less execution time, showing promise for execution on board spacecraft.

Adam P. Herrmann↗

Importance of kernel bandwidth in quantum machine learning

Quantum kernel methods are considered a promising avenue for applying quantum computers to machine learning problems. Identifying hyperparameters controlling the inductive bias of quantum machine learning models is expected to be crucial given the central role hyperparameters play in determining the performance of classical machine learning methods. In this work we introduce the hyperparameter controlling the bandwidth of a quantum kernel and show that it controls the expressivity of the resulting model. We use extensive numerical experiments with multiple quantum kernels and classical data sets to show consistent change in the model behavior from underfitting (bandwidth too large) to overfitting (bandwidth too small), with optimal generalization in between. We draw a connection between the bandwidth of classical and quantum kernels and show analogous behavior in both cases. Furthermore, we show that optimizing the bandwidth can help mitigate the exponential decay of kernel values with qubit count, which is the cause behind recent observations that the performance of quantum kernel methods decreases with qubit count. Here, we reproduce these negative results and show that if the kernel bandwidth is optimized, the performance instead improves with growing qubit count and becomes competitive with the best classical methods.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗